|Publication number||US8111926 B2|
|Application number||US 12/573,889|
|Publication date||Feb 7, 2012|
|Filing date||Oct 6, 2009|
|Priority date||Aug 10, 2009|
|Also published as||US20110033109|
|Publication number||12573889, 573889, US 8111926 B2, US 8111926B2, US-B2-8111926, US8111926 B2, US8111926B2|
|Inventors||Wu-Jie Liao, Ting-Yuan Cheng, Pai-Yu Tien, Hui-Jan Chien, Tsai-Hsing Chen|
|Original Assignee||Primax Electronics Ltd.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (21), Referenced by (1), Classifications (12), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Invention
The present invention relates to an image processing method and an apparatus thereof, and more particularly, to an image processing method for deriving text characteristic images from an input image and a related apparatus.
2. Description of the Prior Art
After scanners capture image data, if text parts of the image data can be detected and printed with pure black ink while, non-text parts are printed with a different printing configuration, the quality and efficiency of the printing process can be greatly improved as well as the readability of the image data after being printed. In addition to using edge detection to capture text parts of an image, the conventional text detection technology also utilizes chroma detection and dot detection to exclude non-text parts. However, the computational cost of the conventional edge detection technology is high, and furthermore, the chroma detection and the dot detection is incapable of locating texts on colored backgrounds or texts nearby image patterns. Therefore, how to capture text parts within an input image swiftly and efficiently has become a research topic in this field.
According to one embodiment of the present invention, an exemplary image processing method for deriving text characteristic parts from an input image is provided. The exemplary image processing method includes: performing a plurality of edge detecting processes upon the input image to generate a plurality of edge images, respectively, wherein the image detecting processes include: performing a first edge detecting process upon the input image to derive a first edge image according to a first upper threshold and a first lower threshold, wherein the first upper threshold is larger than the first lower threshold; and performing a second edge detecting process upon the input image to derive a second edge image according to a second upper threshold and a second lower threshold, wherein the second upper threshold is larger than the second lower threshold, the second upper threshold is smaller than the first upper threshold and the second lower threshold is smaller than the first lower threshold; and deriving a first text characteristic image according to the edge images.
According to another embodiment of the present invention, an exemplary apparatus for deriving text characteristic images from an input image is provided. The exemplary apparatus includes a plurality of edge detection modules and a synthesizing module. The edge detection modules are for performing a plurality of edge detecting processes upon the input image to generate a plurality of edge images, respectively. The edge detection modules include a first edge detection module and a second edge detection module. The first edge detection module performs a first edge detecting process upon the input image to derive a first edge image according to a first upper threshold and a first lower threshold, wherein the first upper threshold is larger than the first lower threshold. The second edge detection module performs a second edge detecting process upon the input image to derive a second edge image according to a second upper threshold and a second lower threshold, wherein the second upper threshold is larger than the second lower threshold, the second upper threshold is smaller than the first upper threshold and the second lower threshold is smaller than the first lower threshold. The synthesizing module derives a first text characteristic image according to the edge images.
These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.
Please refer to
Please refer to
The first edge detection module 103 includes a comparison element 1031, a dilation element 1032 and an outputting element 1033. After receiving the input image IMG, the comparison element 1031 processes each pixel within the input image IMG according to a first upper threshold UT1 and a first lower threshold LT1. If the luminance of the pixel is higher than the first upper threshold UT1, the comparison element 1031 categorizes the pixel as a high-luminance pixel; if the luminance of the pixel is lower than the first lower threshold LT1, the comparison element 1031 categorizes the pixel as a low-luminance pixel. Please note that the first upper threshold UT1 is larger than the first lower threshold LT1 (i.e., UT1>LT1). The dilation element 1032, which is coupled to the comparison element 1031, processes each high-luminance pixel and each low-luminance pixel within the input image IMG. By way of example, the dilation element 1032 categorizes all non-high-luminance pixels distant from the high-luminance pixel by a specific range as high-luminance pixels; and the dilation element 1032 categorizes all non-low-luminance pixels distant from the low-luminance pixel by a specific range as low-luminance pixels. Finally, the outputting element 1033, which is coupled to the dilation element 1032, derives the first edge image IMG_1 according to each pixel categorized as a high-luminance pixel and a low-luminance pixel simultaneously in space.
Please refer to
Please refer to
Please refer to
Please note that, in this embodiment, the image processing apparatus 100 only utilizes the first edge detection module 103 to exclude the background image BG with a luminance B, and utilizes the second edge detection module 104 to exclude the issue of capturing the first text T1 which also has a luminance B. However, the present invention can also utilize different edge detection modules according to different requirements (for example, detecting a text with a specific luminance and a text within a background having a specific luminance) such that a better image processing effect can be achieved. That is, the present invention is not limited to utilizing only two edge detection modules, and other kinds of variation in design, such as using more than two edge detection modules, still obey the spirit of the present invention. For example, the image processing apparatus 100 in
The aforementioned description is only a preferred embodiment of the present invention, where the preprocessing module 102 is for optimizing the quality of the input image IMG and is not necessarily required in a practical implementation of the present invention. Furthermore, as long as the input image IMG contains the luminance information of each pixel, the conversion module 101 is also not necessarily required. In other words, the conversion module 101 and the preprocessing module 102 are optional elements, and if the original image IMG_OG is a processed image itself, the conversion module 101 and/or the preprocessing module 102 can be omitted in other embodiments of the present invention.
In addition, the present invention can also utilize a chroma detection module to detect texts with specific chrominance. Please refer to
To summarize, the present invention provides an image processing method and a related apparatus capable of deriving text characteristic images from an input image. With proper configuration of thresholds, a plurality of edge detection operations can derive a plurality of edge images from an input image. After the edge images are further processed, images of text characteristics can be captured swiftly and efficiently. Additionally, the present invention can operate in conjunction with a chroma detection operation and derive texts on colored backgrounds or texts nearby image patterns.
Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5317420 *||Oct 15, 1992||May 31, 1994||Matsushita Electric Industrial Co., Ltd.||Noise elimination apparatus for eliminating noise components having a high density and a low density|
|US5418862 *||Aug 10, 1992||May 23, 1995||United Parcel Service Of America||Method and apparatus for detecting artifact corners in two-dimensional images|
|US5515480 *||Jun 15, 1994||May 7, 1996||Dp-Tek, Inc.||System and method for enhancing graphic features produced by marking engines|
|US5729632 *||Dec 8, 1994||Mar 17, 1998||Eastman Kodak Company||Reproduction apparatus and method for adjusting rendering with toners of different particle sizes|
|US5732146 *||Apr 18, 1995||Mar 24, 1998||Matsushita Electric Industrial Co., Ltd.||Scene change detecting method for video and movie|
|US5768403 *||Dec 21, 1994||Jun 16, 1998||Fuji Xerox Co., Ltd.||Image-area identifying system for a color image processing apparatus|
|US5818976 *||Dec 20, 1993||Oct 6, 1998||Visioneer, Inc.||Method and apparatus for document skew and size/shape detection|
|US5982911 *||Jun 13, 1996||Nov 9, 1999||Sanyo Electric Co., Ltd.||Braille recognition system|
|US5999647 *||Feb 28, 1996||Dec 7, 1999||Matsushita Electric Industrial Co., Ltd.||Character extraction apparatus for extracting character data from a text image|
|US6064769 *||Nov 5, 1998||May 16, 2000||Nakao; Ichiro||Character extraction apparatus, dictionary production apparatus and character recognition apparatus, using both apparatuses|
|US6176566 *||Mar 10, 1997||Jan 23, 2001||Hewlett-Packard Company||Cell-based text enhancement for a color printer by complete excess-color stripping, based on previous regions|
|US6735338 *||Jun 30, 1999||May 11, 2004||Realnetworks, Inc.||System and method for generating video frames and detecting text|
|US6744919 *||Jul 24, 2001||Jun 1, 2004||Hewlett Packard Development Company, L.P.||Classification of blocks for compression based on number of distinct colors|
|US6917707 *||Sep 28, 2000||Jul 12, 2005||Sharp Kabushiki Kaisha||Image processing method and image processing apparatus|
|US7265873 *||Apr 13, 1999||Sep 4, 2007||Conexant Systems, Inc.||Image processor that performs edge enhancement using a dynamic threshold|
|US7446817 *||Feb 14, 2005||Nov 4, 2008||Samsung Electronics Co., Ltd.||Method and apparatus for detecting text associated with video|
|US7616340 *||Mar 23, 2006||Nov 10, 2009||Fujifilm Corporation||Image processing method and apparatus, threshold value matrix creating method, image forming apparatus, sub-matrix creating method and program|
|US7948655 *||May 24, 2011||Canon Kabushiki Kaisha||Image processing apparatus, image processing method, image processing program, and storage medium|
|US8045809 *||May 22, 2007||Oct 25, 2011||Ricoh Company, Ltd.||Image forming apparatus|
|US20020031263 *||Apr 10, 2001||Mar 14, 2002||Shinji Yamakawa||Method and system for processing character edge area data|
|US20020080377 *||Dec 10, 2001||Jun 27, 2002||Kazunari Tonami||Image-processing device using quantization threshold values produced according to a dither threshold matrix and arranging dot-on pixels in a plural-pixel field according to the dither threshold matirx|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US20140306983 *||Apr 11, 2014||Oct 16, 2014||Samsung Display Co., Ltd.||Data processing device and display system including the same|
|U.S. Classification||382/199, 382/270, 358/3.21, 382/176|
|International Classification||G06K9/48, G06K9/34, H04N1/40|
|Cooperative Classification||G06K9/4604, G06K9/4652, G06K2209/01|
|European Classification||G06K9/46C, G06K9/46A|
|Oct 6, 2009||AS||Assignment|
Owner name: PRIMAX ELECTRONICS LTD., TAIWAN
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LIAO, WU-JIE;CHENG, TING-YUAN;TIEN, PAI-YU;AND OTHERS;REEL/FRAME:023328/0648
Effective date: 20090904
|Sep 18, 2015||REMI||Maintenance fee reminder mailed|
|Feb 7, 2016||LAPS||Lapse for failure to pay maintenance fees|
|Mar 29, 2016||FP||Expired due to failure to pay maintenance fee|
Effective date: 20160207